ECO 391: Exam #3 questions &
answers graded A+ 2025/2026
Homoscedastic - ANS ✔✔Variance of the error term is the same for all observations
Multicollinearity - ANS ✔✔2 or more explanatory variables in a regression are highly linearly
correlated
Which descriptive statistic does not give information about the spread of data?
a) range
b) standard deviation
c) variance
d) mean - ANS ✔✔d
The normal distribution is always symmetric and centered around zero
a) true
b) false - ANS ✔✔b
Taking a sample of Lexington residents will likely result in biased estimates compared to the
entire population of Kentucky.
a) true
b) false - ANS ✔✔a
,The difference between the observed value and the predicted value is called the
a) standard deviation
B) variance
c) error
d) laffer's constant - ANS ✔✔c
Data on income in America tends to have a very long right-tail. Knowing this, which value will be
larger, median household income or mean household income?
a) median household income
b) mean household income
c) they will be equal
d) it will depend on how far out the right-tail goes - ANS ✔✔b
Adding an independent variable, which has no predictive power, to a regression model will
usually lower the value of R^2
a) true
b) false - ANS ✔✔b
Which is a way to increase the value of R^2?
a) drop a variable from the model which is not particularly useful in predicting the dependent
variable
, b) add an independent variable to the model which is useful for predicting the dependent
variable
c) correct for heteroskedasticity
d) reduce multicollinearity by dropping an independent variable - ANS ✔✔b
If we dropped every independent variable except one, what would we expect to happen to the
value of R^2?
a) increase
b) decrease
c) not change
d) not enough info - ANS ✔✔b
For a = 0.05, we will reject the null hypothesis for a p-value that is 0.08
a) true
b) false - ANS ✔✔b
The F-statistic from a regression output is used to test the significance of the entire model
a) true
b) false - ANS ✔✔a
The assumed null hypothesis for regression coefficients is that H0: B=0 is not statistically
significant. Therefore HA; B≠0 is statistically significant
a) true
answers graded A+ 2025/2026
Homoscedastic - ANS ✔✔Variance of the error term is the same for all observations
Multicollinearity - ANS ✔✔2 or more explanatory variables in a regression are highly linearly
correlated
Which descriptive statistic does not give information about the spread of data?
a) range
b) standard deviation
c) variance
d) mean - ANS ✔✔d
The normal distribution is always symmetric and centered around zero
a) true
b) false - ANS ✔✔b
Taking a sample of Lexington residents will likely result in biased estimates compared to the
entire population of Kentucky.
a) true
b) false - ANS ✔✔a
,The difference between the observed value and the predicted value is called the
a) standard deviation
B) variance
c) error
d) laffer's constant - ANS ✔✔c
Data on income in America tends to have a very long right-tail. Knowing this, which value will be
larger, median household income or mean household income?
a) median household income
b) mean household income
c) they will be equal
d) it will depend on how far out the right-tail goes - ANS ✔✔b
Adding an independent variable, which has no predictive power, to a regression model will
usually lower the value of R^2
a) true
b) false - ANS ✔✔b
Which is a way to increase the value of R^2?
a) drop a variable from the model which is not particularly useful in predicting the dependent
variable
, b) add an independent variable to the model which is useful for predicting the dependent
variable
c) correct for heteroskedasticity
d) reduce multicollinearity by dropping an independent variable - ANS ✔✔b
If we dropped every independent variable except one, what would we expect to happen to the
value of R^2?
a) increase
b) decrease
c) not change
d) not enough info - ANS ✔✔b
For a = 0.05, we will reject the null hypothesis for a p-value that is 0.08
a) true
b) false - ANS ✔✔b
The F-statistic from a regression output is used to test the significance of the entire model
a) true
b) false - ANS ✔✔a
The assumed null hypothesis for regression coefficients is that H0: B=0 is not statistically
significant. Therefore HA; B≠0 is statistically significant
a) true
